: Collaborating investigators at the University of Hawaii and Southern California have established a substantial cohort comprised of minorities (African-Americans, Hispanics, Japanese) and Whites. The cohort of 202,136 individuals has been characterized by questionnaires regarding detailed dietary histories, past medical histories and other lifestyle characteristics, and is being followed for all incident cancer cases utilizing population-based cancer registries in each location. This cohort is of sufficient size and heterogeneity to study the relationship between environmental risk factors, as determined primarily by the questionnaire, and genes that may be important alone or as sources of potential interaction in determining the risk of sporadic cancers of the colorectum, prostate, and breast. The proposed design maximizes the efficiency of the cohort by utilizing a nested case-cohort control design. A total of 400 cohort-controls will be randomly selected from each ethnic/sex group and blood will be collected from these cohort controls and from all incident cases of colorectum, prostate, and breast cancer. Specifically, this proposal will investigate polymorphisms in candidate genes (NAT2, GSTM1, CYP1A1, SRD5A2, AR, EDH17B2, CYP17, and ER), that might alter metabolism of relevant endogenous or exogenous exposures and their interactions with dietary variables (e.g., saturated fat and fiber intake) and other exogenous exposures (e.g., hormone replacement therapy) that affect the risk of these three cancers. In addition to case control comparisons, the design allows for comparison of the frequency of these genetic polymorphisms across ethnic groups.